(+)-morphine - American Chemical Society

Aug 16, 1977 - Studies in the ( + )-Morphinan Series. 4.1. A Markedly ImprovedSynthesis of (+)-Morphine. Ikuo Iijima,23 Jun-ichi Minamikawa,2b Arthur ...
0 downloads 0 Views 323KB Size
1462 J . Org. Chem., Vol. 43, No. 7, 1978

Notes

(7) While one could picture iminotriphenylphosphorane itself effecting the

transformation described here, its experimental use proves to be less than ideal. Since simple exposure to the atmosphere results in hydrolysis to ammonia and triphenylphosphine oxide, rigorously anhydrous conditions are required for its synthesis, storage, and use. Reactions of this reagent with other than the most highly reactive of ketones are thus often complicated by decomposition of the reagent. In contrast, the trimethyisilyl-protected compound employed here is stable to atmospheric moisture and requires no undue care in its manipulations.

Studies in t h e (+)-Morphinan Series. 4.' A Markedly Improved Synthesis of (+)-Morphine Ikuo Iijima,2aJun-ichi Minamikawa,2bArthur E. Jacobson, Arnold Brossi and Kenner C. Rice* Medicinal Che,nistry Section, ATational Institute of Arthritis, Metabolism, and DigestiLse Diseases, National Institutes of Health, Rethesda, Maryland 20014 Received September 16,1977

The absolute configuration of the morphinan skeleton of (-)-sinomenine (1)3is enantiomeric to natural (-)-morphine, and conversion of 1 into (+)-morphine (IO) was reported by ,CH,

5 '

N

0 (XO

OCH,,

OH

CH,O

OH

2a, R , = OCH,; R, = H b, R, = H; R, = OCH,

1

CH,

,CH3 N

N *'

CH

,CH;

R,

CH,O

OCH, 5

3, R , , R, = 0; R, = H 4 , R, = R, = OCH,; R, = H 6, R, = R, = OCH,; R, = Br

,CH3 Y

7 , R , = R, = OCH,; R, = CH,

8, R , , R , = 0 ; R, = CH, 9, R , = OH; R, = H; R, = CH, 1 0 , R , = OH; R, = R, = H 11,R , = OAc; R, = H ; R, = Ac

Goto's group.4a-cTo define morphine's interaction with opiate receptors more clearly,5we wanted to prepare large quantities of unnatural enantiomer 10 and several of its congeners. We now summarize our results that followed, in principal, Goto's original s ~ h e m e , but ~ ~ which - ~ implemented novel reactions and the findings of others, especially those of Rapoport et a1.6 Rapoport's results in the natural (-) series became available only after we had started our project. A tenfold increase in the overall yield of (+)-morphine ( l o ) , previously reported by

Goto, from (-)-sinomenine (1) was accomplished as follows. Catalytic reduction7 of 1 afforded a mixture of two diastereomers (2a and 2b) which was separated by preparative thin-layer chromatography. The equilibrium mixture was reestablished by brief boiling of either isomer in methanol. Deuterium exchange of 2a and 2b allowed definitive assignment of the chemical shift of the C-5 and C-7 protons. Assuming that the preferential conformation of ring C is the chair form, the absolute configuration of 2a and 2b at C-7 could be determined. Major isomer 2a and minor isomer 2b were assigned 7R (axial H ) and 7 s (equatorial H) configurations, respectively, for the following reasons. The chemical shift of the C-7 equatorial proton, due to the shielding effect of the carbonyl group in the 7 s isomer, lies upfield of the C-7 axial proton in the 7R isomer, in accord with previous work on a-methoxydecaIones,* The chemical shift of the C-7 proton of the 7 s isomer was 6 3.36 (t,J = 3.5 Hz) and that of the C-7 proton in the 7R isomer was 6 3.90 (center of d of d, J = 7 , 1 2 Hz), and the coupling constants were of the magnitude expected. The equatorially oriented C-7 methoxyl group in the 7R isomer was deshielded by the carbonyl group (6 3.43), as compared with the methoxy group in the 7s isomer (6 3.301, again in accord with cu-methoxydecalones.hMolecular models (Dreiding) indicate that the major product might well be the 7R isomer because of the less sterically hindered methoxyl group. The C - 5 equatorial proton in both 7 R and 7s isomers was considerably deshielded, presumably due to its proximity to the aromatic ring (see Experimental Section). A similar effect was noted in dihydrothebainone.6 Since the next step, the acid-catalyzed SN~' cyclization of 2a and 2b to 3, with loss of methanol, proceeds under conditions which equilibrate the two epimers. the mixture was treated directly with polyphosphoric acid a t 65-70 "C. Ketone 3 is rather stable under these reaction conditions, in contrast to Goto'sg more drastic conditions; and yields of desired ketone 3 were consistently 70-75%. Introduction of a double bond in the 7,8 position of 3 is not easy, and attempts to introduce it by direct oxidation were unsuccessful. This could, however, be accomplished by phenylselenation and oxidative elimination. but only after the N-methyl group was replaced by a N-carbethoxy group.1° Meanwhile, Rapoport's modification for converting (-)dihydrocodeinone (enantiomer of 3) into (-)-codeinone (enantiomer of 8) became known6 and was successfully implemented in our plan, which now took the following course: ketalization of 3 to dimethyl ketal 4 (98%); elimination of methanol with p -toluenesulfonic acid in chloroform to give enol methyl ether 5 (83%); addition of methyl hypobromite, leading to bromodimethyl ketal 6 (75%);elimination of HBr with potassium tert- butoxide in MezSO at room temperature instead of 60 "C6 to give 7 (87%);and deketalization of 7 with 5% HC1 instead of AcOH6 to give (+)-codeinone (8; 96%). Compounds 3 and 5-8 showed the properties previously reported by Goto et al., and 3-8 had properties identical with the corresponding compounds in the (-) series prepared as described by Rapoport,6 except for the optical rotation. Reduction of unsaturated ketone 8 with sodium borohydride in methanolll afforded (+)-codeine (9),which was converted into (+)-morphine (10) by O-demethylationl* with boron tribromide in chloroform. Unknown (+)-heroin ( 1 1 ) was obtained from 10 by treatment with acetic anhydride. Crystallization from ethyl acetate gave prisms identical with authentic (-)heroin (enantiomer of 1 l ) ,except for the sign of optical rotation. (-)-Heroin showed specific optical rotation 10" higher than previously reported.I3 (+)-Codeine (91, (+)-morphine (lo), and (+)-heroin (11) showed no analgesic activity on subcutaneous injection in mice in routine screening for centrally active analgesics. Unnatural

This article not subject to U S . Copyright. Published 1978 by the American Chemical Society

J . Org. Chem., Vol. 43, No. 7,1978

Notes (+)-morphine (lo) showed interesting central effects when injected intracerebrally in rats, suggesting the existence of multiple morphine receptors in the brain.I4

Experimental Section Melting points were determined on a Thomas-Hoover melting point apparatus and are corrected. Elemental analyses were performed by the Section on Microanalytical Services and Instrumentation of this Laboratory. The identity and chemical purity of 3-11 were confirmed by direct comparison with authentic samples of the enantiomeric (-1 series. IR, NMR (using tetramethylsilane a t 6 0.0 as an internal reference), and mass spectra were obtained on Perkin-Elmer 257, Varian Model HR-220, and Hitachi RMU-6E (70 eV) instruments, respectively. Optical rotations were measured with a Perkin-Elmer Model 141 polarimeter. Silica gel G F plates for analytical and preparative TLC were purchased from Analtech, Inc., Newark, Del. 7 ( R ) -and 7( S)-(+)-Dihydrosinomenine(2a and 2b). Sinomenine (1; 16.6 g, 50.40 mmol) was dissolved in MeOH (450 mL) and hydrogenated using 10% Pd/C as a catalyst until the absorption of hydrogen stopped a t approximately 1 mol. After filtration of the catalyst, the solvent was evaporated t,o give an oil residue which solidified on the addition of ether (200 mL), and the product was collected by filtration. This product (16.5 g) melted a t 190-195 "C (lit.7 m p 198 "C) and was used for conversion to 3 without further purification. Preparative TLC of a portion (150 mg) of this mixture of position 7epimers over silica gel GF (EtzO-MeOH, 91) gave as the major component the lower R f 7R epimer (100 mg, 67%),which showed, after crystallization from CHC13-Et20 (l:lO), mp 196.5-197.5 " c ; [aIz3D +121" ( e 1.28, CHC13);IR (CHC13) 1733 cm-'; NMR (CDC13) 6 6.60 ( 2 H , AB system, J = 8 Hz, aromatic H), 6.44 (1 H, brd s, OH), 4.30 (1 H , d , J = 13 Hz, C-5 equatorial H), 3.90 (1H , dd, J = 7,12 Hz, C-7 axial H), 3.80 (3 H , s, aromatic OCH3),3.43 (3 H , s, C-7 OCH3),2.41 (3 H , s, N-CH3). 2.25 (1 H, d ?J = 13 Hz, C-5 axial H). Anal. Calcd for CIgH25N04: C, 68.86; H , 7.60; N, 4.23. Found: C, 69.04; H, 7.54; K!4.18. The higher R, 7s epimer (39 mg, 26%) was obtained as the minor isomer and showed, after crystallization from CHC13-Et20 (l:lO), mp 196.5-197.5 "c;[ a I z 3+87" ~ (C 1.49, CHC13);IR (CHC13) 1725 cm-'; NMR (CDC13)6 6.64 (2 H , dd, J = 8 Hz, aromatic H), 6.26 (1H , brd s: OH), 4.07 (1 H,d , J = 13 Hz, C-5 equatorial H ) , 3.82 (3 H, s, aromatic OCH3),3.36 (1H , t, J = 3.5 Hz, C-7 equatorial H ) , 3.30 (3 H , s, C-7OCH3),2.73 (1H , d , J = 13 Hz, C-5 axial H), 2.42 (3 H, s, NCH!I). Anal. Calcd for C19H2jN04: C, 68.86; H , 7.60; N, 4.23. Found: C, 69.08; H , 7.54; N, 4.13. (+)-Dihydrocodeinone (3). The mixture of 2a and 2b from above (3.00 g, 9.05 mmol) and polyphosphoric acid (Matheson, Coleman and Bell, 60 g) was heated at 65-70 "C for 1.25 h while stirring. The cooled reaction mixture was basified by the careful addition of ammonium hydroxide (28%)at 0 "C, saturated with NaCl, and extracted with CHC13. The extracts were dried over Na2SO4and evaporated to afford a solid, which was recrystallized from EtzO-CHC13 (3:l) to give 3 (1.98 g, 7%): m p 196 5-197.5 "C (lit.gmp 197-198 'C); [aIz3D +205.3" (c 0.9, CHC13) [lit., ["ID f207.4" (CHC13)I. Conversion of (+)-Dihydrocodeinone (3) to (+)-Codeine (9). (+)-Codeine (9 I was prepared from (+)-dihydrocodeinone (3) essentially as described by Rapoport'j in the (-1 series via the following intermediates. (+)-Dihydrocodeinone dimethyl ketal 4: mp 121-122 " C ; ['2Iz3D +167.2" (c 1.1, EtOH) [lit.'5 (-) enantiomer of 4: mp ~ (c 0.9, EtOH)]. (+)-8,14-Dihydrothebaine 122-123 "C, [ a ]-151' ( 5 ) : m p 162-16:; "c;[CYIz3D +268.6" ( C 1.1, C'jH6) [lit.4cmp 162-163 "c,["ID +268.2" (c 1.544, C & , ) ] . (+)-7-Bromodihydrocodeinone dimethyl ketal 6: mp 116-117 "c;[aIz3D +165.1" (c 1,CHC13) [lit.4c ~ (c 1.536, CHCls)].Treatment of 6 with pom p 117 "C, [ a ]+164.5" tassium tert-butoxide at 25 "C (48 h) instead of 60 "C6 gave (+)codeinone dimethyl ketal 7: mp 135-136.5 "C; [ m I 2 3 +238.3" ~ (c 1.2, EtOH) [lit.4cmp 138 "C, [ a ] D +236O ( e 1.016, EtOH)]. Hydrolysis of 7 with SohHCl(0.5 h. 75 "C) instead of AcOH-H206 gave (+)-codeinone 18): m p 185-186 "c;[aIz3D +204.5" (c 1, EtOH) [lit.4dm p 186 "C, [N]D +2O6.Oc' (EtOH)].Reduction of 8 with NaBH4 in MeOH gave (+)-codeine (9) mp 157.5-158.5 "C; [aIz3D f136.2" (c 0.7, EtOH) [lit.4am p 158 '(:, [CY]~'D +137.4' (c 0.743, EtOH)]. (+)-Morphine (10). To a stirred solution of BBr3 (6.00 g, 24 mmol) in CHC13 (70 ml,) was added 9 (1.167 g, 3.9 mmol) in CHC13 (10mL) at23-26 "C over a 2-min period, and stirring was continued for 15 min. The reaction mixture was poured into a stirred mixture of ice (32 g) and NH40H (28% NH3,8 mL) and stirred for 30 min a t 0 "C. The crystalline material which formed was filtered, washed with cold CHCll and then water. and dried to give 10 (800 mg). The aqueous

1463

phase from the filtrate was saturated with NaCl and extracted with CHC13-EtOH (3:l).The combined extracts were evaporated, and the residue was purified by silica gel thin-layer chromatography using CHC13-MeOH (8:2) as a solvent to yield 10 (241 mg). Total yield was 88%.Recrystallization from MeOH gave 10 .HzO as colorless prisms: ~ (c 0.49, MeOH) [lit.4amp247-248 "C, mp 253-255 "C; [ a I z 3t132.1' (c 0.383, MeOH)]. [ a ] 2 3t132.1' ~ Anal. Calcd for C17H19N03"20: C, 67.30: H , 6.98; N! 4.62. Found: C, 67.47; H, 7.25; N, 4.63. (+)-Heroin (11).A mixture of 10 (285 mg, 1 mmol) and acetic anhydride (2 mL) was heated a t 90-100 "C for 4 h. Ether was added to the cooled solution, and the mixture was basified with 1oo/o KOH while cooling. The ether phase was separated, the aqueous phase was extracted with ether, and the combined extracts were dried over Na2S04. The solvent was evaporated to give a solid, which was recrystallized from AcOEt to afford 11 (295 mg, 80%): mp 169-170.5 "C; [cY]~~D +176" (c 0.63, MeOH) [lit.13(-1 enantiomer of 11: mp 173 "C, [ a I z 5 ~ -166.4' ( e 1.49, MeOH)]. Anal. Calcd for C21H23N05: C, 68.28; H , 6.28; N. 3.79. Found: C. 67.97; H, 6.37; N, 3.44.

Acknowledgment. We would like to thank Drs. Mario Aceto, Louis Harris, and Everette L. May, Medical College of Virginia, for their encouragment. The analgesic screening of the compounds mentioned was carried out by Mrs. Louise Atwell of our laboratory. We also thank Dr. Mikio Takeda, Tanabe Seiyaku Research Laboratory, and Dr. Wataru Nagata, Shionogi Seiyaku Research Laboratory, for the generous gifts of (-)-sinomenine used in this work. Registry No.-1, 115-53-7; 2a, 65120-75-4; 2b, 65120-76-5; 3, 64520-24-7; 4,65165-95-9; 5,65165-96-0; 6,65165-97-1; 7,65165-98-2; 8,65494-91-9; 9,64520-25-8; 10,65165-99-3; 11,65166-00-9.

References and Notes (1) I. iijima, K. C. Rice, and A. Brossi, Helv. Chim. Acta, 60, 2135 (1977). (2) (a)Visiting Scientist, NIAMDD, 1975-1977. (b) Visiting Scientist, NIAMDD, 1976-1978. (3) H. L. Holmes, Alkaloids (N.Y , ) , 2, 219 (1952). (4) (a)K. Goto and I. Yamamoto, Proc. Jpn. Acad., 30, 769 (1954);(b) 33, 477 (1957);(c) 34, 60 (1958);(d) K. Goto, "Sinomenine, an Optical Antipode of Morphine Alkaloids", Kitasato Institute, Tokyo, 1964, p 96. (5) A. H. Beckett and A. F. Casy, J. Pharm. Pharmacol., 6, 986 (1954);P. S. Portoghese, J. Pharm. Sci., 55, 865 (1966);A. Goldstein,L. I. Lowney, and B. K. Pol, Proc. Natl. Acad. Sci. U.S.A., 68, 1742 (1971); C. B. Pert and S. H. Snyder, Science, 179, 1011 (1973); E. J. Simon, J. M. Hiller, and I. Edelman, Proc. Natl. Acad. Sci. U.S.A., 70, 194 1 (1 973):L. Terenius. Acta Pharmacol. Toxicol., 33, 377 (1973). (6) D. D. Weller and H. Rapoport, J. Med. Chem. 19, 1171, (1976). (7) K. Goto and S. Mitsui, Bull. Chem. SOC.Jpn., 5, 282 (1930). (8) L. M. Jackman and S. Sternhell in "Applications of Nuclear Magnetic Resonance Spectroscopy in Organic Chemistry",2nd ed, Pergamon Press, Elmsford, N.Y., 1969, p 241. (9) K. Goto and 1. Yamamoto, Roc. Jpn. Acad., 36, 145 (1960); 34, 60 (1958). (10) I. lijima, K. C. Rice, and J. V. Silverton, Heterocycles, 6, 1157 (1977) (11) M. Gates, J. Am. Chem. Soc,, 75, 4340 (1953). (12) K. C. Rice, J Med. Chem., 20, 164 (1977). (13) The Merck Index 1976, No. 2925: [.Iz5D -166" ( c 1.49, CH30H).Our

rigorously purified sample of (-)-heroin showed [ L Y ] * ~-174.7" ~ ( c 1.15, MeOH) after recrystallization from ethyl acetate and drying. A satisfactory combustion analysis was obtained on this sample. (14) Y. F. Jacquet, W. A. Klee, K. C. Rice, I. lijima, and J. Minamikawa, Science, 198, 842 (1977).

(15) H. Rapoport, H. R. Reist, and C. H. Lovell, J Am. Chem. SOC.,78,5126 (1956).

Cleavage of Tetrahydrofuran by Lithium Bis(2,6-di-tert-buty1phenoxy)aluminumHydride H. Haubenstock* and N.-L. Yang

The City University of New York, The College of S t a t e n Island, S t a t e n Island, N e u York 10301 Receiued August 16, 1977

The reaction of lithium aluminum hydride (LiAlH4)with 2 or 3 molar equiv of 2,6-di-tert-butylphenol (I) a t room temperature gives lithium bis(2,6-di-tert-buty1phenoxy)aluminum hydride (11; eq 1).This is confirmed in Table I for

0022-326317811943-1463$01,00/00 1978 American Chemical Society